Don’t worry, predictions from orgs directly profiting from hyping bullshit with zero consequences always pan out.

Like many innovations that came before, LLMs and GenAI are best used as tools to assist those whose are already experts in their field. This reminds me of an old adage of mine:

If you give a Ferrari to someone who can't drive, you don't get faster journeys; you get messier accidents.

I'm a fan of good documentation, but I also recognize its limits.

Using an adblocker in 2025 is a security must have, the amount of investment, software and product scams is crazy

In the Aurora DSQL preview I see very high disconnect duration outliers using the Lambda example code form the docs

Seeing millions of GET API requests with rarely changing data by a single customer makes me wonder if this is a symptom of a "too fast API", never creating the user thought that caching would be a good idea

IT'S HERE! Amazon Aurora Distributed SQL (DSQL). It's DynamoDB and Postgres merged into a truly serverless relational database.

This is THE announcement of Re:Invent 2024. Everything else is irrelevant.

Doing a seamingly small change on a node.js project. Prettier having strange problems, then jest is broken and after fixing both, webpack crashes with a quite generic error.
You have to love it

Das ist eine so fürchterliche, eine so bescheuerte Entscheidung. Wer jemals einen leidenden Menschen gesehen hat, der ohne Aussicht auf Heilung einfach in Ruhe sterben möchte, weiß das.
Was für eine Würde nach GG Art 1 soll das sein?

Generating random numbers in C

It took me way too long to find this Terraform import block error. Terraform only gave me a cryptic error message.

The problem, the "to" string was quoted

Tiefststand bei der Stromerzeugung aus Kohle!
Im dritten Quartal 2023 wurden 22,2 TWh Strom aus Kohle in Deutschland produziert. Das ist der bisher niedrigste Wert in den mir zur Verfügung stehenden Daten seit 1990.
Hat jemand noch ältere Daten?
energy-charts.info/charts/energ...

Love a good metadata error